BTS 1.0: A Computer Package to do a Complete Bayesian Analysis of Autoregressive Processes | ||
| The Egyptian Statistical Journal | ||
| Article 5, Volume 59, Issue 2, December 2015, Pages 213-236 PDF (30.91 M) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/esju.2015.314466 | ||
| Authors | ||
| Bakri M.Assas1; Samir M.Shaarawy2; Emad E.A. Soliman2; Sherif S.Ali2 | ||
| 1umm A Qura University | ||
| 2Cairo University | ||
| Abstract | ||
| The proposed work intends to assert a menu driven computer package to conduct a complete Bayesian analysis of autoregressive models. The analysis includes the four well known steps of time series analysis, namely, identification, estimation, diagnostic check and prediction. The proposed package conducts the above mentioned four steps using the Bayesian approach in time series analysis. Furthermore, it checks the stationarity of the time series and transforms the non-stationary time series into a stationary one before performing the steps of the analysis. The package depends on a theoretical Bayesian framework published by the authors in well-known journals. The package includes six modules to perform the complete Bayesian analysis for autoregressive models. It is easy to use the package by specialists and non-specialists in modeling and analyzing real data time series that have an evidence to follow an autoregressive process in their behavior. | ||
| Keywords | ||
| Time Series Analysis; Bayesian Analysis; Autoregressive Models | ||
|
Statistics Article View: 46 PDF Download: 1 |
||
